First construction of a saricandin analog corresponding to papulacandin D

Article abstract of DOI:10.1055/s-2001-13399

The first total synthesis of a saricandin analog corresponding to papulacandin D has been achieved via a highly convergent synthetic strategy. A readily accessible chiral building block 3 was designed and prepared in large scale via an enantioselective re

Full text of DOI:10.1055/s-2001-13399

PAPER
833
First Construction of a Saricandin Analog Corresponding to Papulacandin D
F
irst Construction
h
of
a
Sarican
a
din
A
nalog
f
Corresp
i
onding
q
to
P
apulacandin DHamdouchi,*a Concha Sanchez-Martinezb
a
Lilly Research Laboratories, A division of Eli Lilly and Company, Lilly Corporate Center, DC: 0540, Indianapolis, IN 46285; USA
b
Centro de Investigación Lilly, Avenida de la Industria 30, 28108-Alcobendas, Madrid, Spain
E-mail: Hamdouchi_Chafiq@Lilly.com
Received 11 September 2000; revised 4 December 2000
Recently, a new potent antifungal antibiotic produced by
the fungal culture, AB 2202W-161, named saricandin was
isolated and characterized.4 Guided by the hypothesis that
saricandin may be formed following an analogous biosyn-
Abstract: The first total synthesis of a saricandin analog corre-
sponding to papulacandin D has been achieved via a highly conver-
gent synthetic strategy. A readily accessible chiral building block 3
was designed and prepared in large scale via an enantioselective re-
duction with pinanyl-9-BBN. The adaptability of compound 3 to- thetic pathway to the papulacandins, we postulated that
ward structural modifications and the highly convergent nature of
compound 1 (Figure) may be an intermediate in the pro-
cess.5 Herein we wish to describe the first total synthesis
the approach is illustrated in the construction of the side chain
present in saricandin by Pd-catalyzed cross-coupling of 2 and 3 and
of compound 1. By analogy with the relationship between
sequences that include triple bond reduction of fragment C(5–16)
papulacandin D and A, compound 1 would address the
and generation of the double bond (C4-C5) using Horner-Emmons
question of how the biological profile of saricandin is af-
fected by the absence of the galactose moiety together
with the short fatty acid. Furthermore, the significance of
such a synthesis is that it may provide a new starting point
for the design of new antifungal agents.
reaction. The assembly of the spirocyclic monoglycoside with sari-
candin side chain is described. A practical technique for isolating
the final product 1 after deprotection with TBAF is discussed. Com-
pound 1 was evaluated for its antifungal activity in enzyme assay
and cell based assays. However, in contrast the activity reported by
Traxler for papulacandin D, the presence of the galactose moiety to-
gether with the short fatty acid in natural saricandin seem to be es-
sential for the antifungal activity.
Our retrosynthetic analysis of compound 1 is outlined in
Scheme 1. We envisioned that disconnection of the ester
would lead to the vinyl iodide 2, the propargylic alcohol 3
along with the Horner-Emmons reagent and the spirocy-
clic glycoside 4. Its ready accessibility and its adaptability
toward structural modifications for possible structure ac-
tivity relationship studies dictated the selection of the
chiral building block 3. The highly convergent nature of
the approach is illustrated in the construction of the side
chain via Pd-catalyzed cross-coupling of 2 and 3 and se-
quences that include triple bond reduction of fragment
C(5–16) and generation of the double bond (C4-C5) via
Horner-Emmons reaction.
Key words: Saricandin, papulacandin, antifungal, enatioselective
reduction
In 1977, Traxler and coworkers reported the isolation and
characterization of a family of antifungal antibiotics
named papulacandins A, B, C and D from Papularia
sphaerospema.1 The compounds displayed potent in vitro
activity against Candida albicans and various other yeasts
and have been shown to inhibit -1,3-glucan synthase, an
enzyme involved in fungal cell wall biosynthesis. Howev-
er, little or no efficacy has been found in animal models.
The papulacandins A, B and C contain a spirocyclic dig-
lycoside and two unsaturated fatty acids, linked as esters
to two hydroxyl groups of the diglycosides. Papulacandin
D, a monosaccharide relative is the simplest member of
the family. The antifungal properties of papulacandin D
juxtaposed with lack of access to natural material and the
synthetically challenging structural feature make this
compound interesting synthetic target.2 The first total syn-
thesis and full stereochemical assignment of papulacandin
D was reported by Barrett and coworkers3 in 1996. Al-
though there is no evidence for a biosynthetic pathway it
is possible that papulacandin D, lacking the galactose res-
idue and the short fatty acid chain (Figure 1), is formed as
intermediate in the biosynthesis of papulacandins A, B
and C.
1
OH
10
5
O
P
O
EtO
EtO
9
4
I
OTIPS
OEt
2
3
9
OTIPS
TIPSO
O
O
O
t-Bu
Si
O
OH
t-Bu
OH
4
Scheme 1
Synthesis 2001, No. 6, 04 05 2001. Article Identifier:
1437-210X,E;2001,0,06,0833,0840,ftx,en;M02200SS.pdf.
© Georg Thieme Verlag Stuttgart · New York
ISSN 0039-7881
The asymmetric synthesis of the chiral building block 3
begins with the protection of propane-1,3-diol as the

834
C. Hamdouchi, C. Sanchez-Martinez
PAPER
OH
O
OH
O
HO
HO
O
O
O
O
R
HO
HO
O
Biosynthetic pathway
O
O
OH
HO
OH
O
O
HO
OH
OH
OH
OH
O
Papulacandin D
Papulacandin A R =
Papulacandin B R =
Papulacandin C R =
OH
OH
O
O
OH
OH
O
HO
HO
O
O
O
O
R
HO
HO
O
O
O
O
OH
O
HO
OH
Biosynthetic pathway?
O
HO
OH
OH
OH
OH
1
O
Saricandin R =
Figure Structures of papulacandin A-D, Saricandin and the related compound 1
mono-TIPS ether 5. Subsequent oxidation with PCC fol-
lowed by condensation of the resulting aldehyde 6 (see
experimental) with lithio(trimethylsilyl) acetylene fur-
1) PCC, CH2Cl2
(93%)
TIPSCl, Im, DMF
(67%)
OH OTIPS
OH OH
2) TMS-acetylene
n-BuLi, THF
( 74%)
nished the racemic 76 in 72% yield. Oxidation of the sec-
ondary alcohol using PCC produced the , -acetylenic
ketone 8, which was then subjected to an asymmetric re-
duction. Enantioselective reduction with pinanyl-9-BBN
using Midland’s conditions7 proved to be very efficient
and allowed the preparation of (S)-( )-7 on large scale in
5
O
OH
PCC, CH2Cl2
(71%)
OTIPS
OTIPS
TMS
good yield and in 97% ee as determined from the Mosher
esters.
TMS
(+)-7
8
The C-silyl group was removed from 7 under conditions
that retained the TIPS group protecting the alcohol. Thus
treatment of (S)-( )-7 with AgNO3 in ethanol-water fol-
lowed by quenching with a solution of KCN afforded 3 in
92% yield. The coupling of the propargylic alcohol 3 with
vinyl iodide (E)-28 was performed with catalytic amount
of palladium(0) under standard protocol9 to afford the (E)-
OH
(–)-Pinanyl-9 BBN, THF
( 75%)
OTIPS
TMS
S-(–)-7
97% ee
Scheme 2
Synthesis 2001, No. 6, 833–840 ISSN 0039-7881 © Thieme Stuttgart · New York

PAPER
First Construction of a Saricandin Analog Corresponding to Papulacandin D
835
enyne alcohol 10 in 79% yield. When compound 10 was um perruthenate (TPAP), introduced by Ley14 for the ox-
subjected to the conditions for the triple bond reduction idation of alcohols proved successful. Thus, treatment of
with LAH,10 we were pleased to find that in addition to the alcohol 14 with catalytic amount of TPAP in the pres-
generating the expected (E,E)-diene, removal of TIPS ence of NMO furnished smoothly the aldehyde 15 in 68%
ether leading directly to the desired diene 11 in 82% yield yield. The reaction was performed on small and large
also resulted. Similar O-silyl bond cleavage affected by scale and the desired aldehyde was isolated in pure form
LAH has been observed previously and attributed to a hy- by simple filtration. With the aldehyde 15 in hand, the
dride transfer from the chelated O-aluminum hydride.11
next step was the generation of the second (E,E)-diene and
completion of the side chain. The aldehyde 15 was sub-
mitted to Horner Emmons reaction using triethyl
phosphonocrotonate (9). The Horner-Emmons reaction
was studied using LiHMDS, KHMDS and NaH. The best
result with regard to the stereoselectivity and the yield of
the reaction was achieved using the t-BuOLi, conditions
OH
OH
AgNO3, KCN
OTIPS
OTIPS
EtOH, H2O
(92%)
TMS
H
S-(–)-7
S-(–)-3
1
described recently by Posner. H NMR analysis of the
I
crude reaction mixture showed an 8:1 mixture of E,E- and
E,Z-isomers. A simple column chromatography afforded
the desired isomer (S)-(+)-16 in 74% yield.
OH
2
OTIPS
Pd(PPh3)4
PhH, (i-Pr)2NH, CuI
10
(79%)
For the synthesis of the tricyclic spiroketal nucleus 4 we
used reaction sequences similar to that described by
Barrett3 for the synthesis of papulacandin D. The process
involved a condensation of aryllithium reagent with pro-
tected gluconolactone and subsequent acid-catalyzed spi-
rocyclization.15
OH
LiAlH4, THF
(82%)
OH
11
OTES
1) Piv-Cl, DMAP, –78 °C
OTES
O
KOSiMe3, THF
(91%)
OH
2) TESCl, Imidazole; DMAP, 23 °C
3) DIBAL, THF, –78 °C
(59%, three steps)
OEt
14
16
Cl
COCl
Pr4NRuO4 (TPAP)
NMO, CH2Cl2
OTES
1)
Cl
CHO
Cl
DMAP, Et3N, THF
(68%)
OTES
O
15
OH
17
2)
O
4, DMAP, DMF
EtO
EtO
P
CO2Et
OTES
9
CO2Et
OTIPS
t-BuOLi, THF
(74%)
S-(+)-16
(E,E):(E,Z) 8:1
OTIPS
TIPSO
O
O
TIPSO
O
Si
O
O
O
O
OH
O
O
O
Scheme 3
O
Si
O
OH
Selective protection of primary alcohol of 11 as its piv-
aloate at 78 ºC followed by silylation of the secondary
alcohol as the silyl ether 13 (see experimental), then re-
duction of the pivaloate ester furnished the primary alco-
hol 14 in 59% overall yield. The oxidation of the primary
alcohol to aldehyde 15 initially proved problematic. The
use of PCC under variety of conditions was found to be Scheme 4
sluggish and afforded a very low yield of 15. Swern oxi-
dation or Dess Martin12 oxidation provided mainly the
product of elimination. Our efforts to circumvent this
problem by applying the combination of Swern oxidation-
Wittig condensation procedure reported by Ireland and
coworkers13 for highly unstable aldehyde was likewise
unsuccessful. Ultimately, the use of tetrapropylammoni-
OTES
OTES
C-2 18
C-3 18
Next, the unsaturated ester 16 was smoothly dealkylated
to the corresponding acid using potassium trimethylsil-
anolate. The acid was then reacted with 2,4,6-trichlo-
robenzoyl chloride in the presence of DMAP. The
resulting mixed anhydride was treated with the spiroketal
Synthesis 2001, No. 6, 833–840 ISSN 0039-7881 © Thieme Stuttgart · New York

836
C. Hamdouchi, C. Sanchez-Martinez
PAPER
diol 4 in the presence of DMAP in DMF to afford a 3:1 In summary, we have described in this paper the first total
mixture of the C-3 and C-2 esters. After separation and synthesis of an analog of Saricandin corresponding to
purification by column chromatography the desired C-3 papulacandin D. We have demonstrated that in contrast to
ester 18 was isolated in 31% yield.
the papulacandin D originally isolated and tested by Trax-
ler and found to be active, the presence of the galactose
moiety together with the short fatty acid in natural sari-
candin are essential for the antifungal activity. The ready
accessibility of the chiral building block used in this syn-
thesis along with its adaptability toward structural modifi-
cations and the highly convergent nature of the approach
promised its use in structure activity relationship studies.
OH
HO
O
O
HO
HO
O
1) TBAF, THF
2) Column chromatography
(MeOH–ACN–CH2Cl2, 5:40:55)
OH
O
C-3 18
All reagents were purchased from Aldrich and used without further
purification. Solvents were purified by distillation: CH2Cl2 (CaH2,
N2), THF and Et2O (Na, benzophenone, N2), CCl4 (CaCl2), benzene
(Na, benzophenone, N2), i-Pr2NH (CaH2, N2) and DMF (24 h over
4 Å molecular sieves). Column chromatography was carried out on
flash silica gel (Merck 230–400 mesh). TLC analysis was conduct-
ed on silica gel plates Whatman. 1H and 13C spectra were recorded
on a 200 MHz Bruker AC-200P or a 300 MHz Bruker AMX, with
CD3OD or CDCl3 as solvents (without internal reference). 1H NMR
data were assigned by irradiation experiments. 13C NMR data were
assigned by DEPT experiments. Mass spectra were recorded on
VG-Autospec mass spectrometer. In the case of the FAB technique
m-NBA was the matrix for low resolution and PEG for high resolu-
tion (HRMS). Optical rotations were measured with a Perkin Elmer
343 polarimeter.
TASF, THF
OH
OH
1
HO
O
HO
HO
O
OH
O
O
(E)-1-Iodohept-1-ene (2)
A flame-dried 500 mL round-bottom flask equipped with a reflux
condenser was purged with N2. Hept-1-yne (19.23 g, 200 mmol)
and anhyd pentane (40 mL) were introduced via syringe, then
DIBAL-H (1 M in toluene, 200 mL) was added dropwise via can-
nula very slowly (1 h). After the addition was completed, the solu-
tion was heated at 100 °C for 3 h. The solvents were removed after
cooling to r.t. and anhyd THF (80 mL) was added. The mixture was
cooled to 50 °C and I2 (50.76 g, 200 mmol) dissolved in anhyd
THF (80 mL) was added dropwise under N2. The reaction mixture
turned brown and was allowed to warm to r.t. (outside the bath),
then hydrolyzed very slowly with H2SO4 (10%, 160 mL) in an ice-
bath. This mixture was poured into another mixture of H2SO4 (160
mL)-ice, and extracted with pentane (2 200 mL). The combined
organic layers were washed with aq sat. Na2S2O3 solution (50 mL),
aq sat. NaHCO3 solution and dried (MgSO4). Solvents were re-
moved and the product was distilled under reduced pressure (70 °C/
8 Torr) in the presence of copper to give 24.7 g (5%) of a colorless
liquid.8
19
Scheme 5
The final deprotection step was next addressed. When us-
ing TBAF desilylation, Barrett et al.3 reported difficulties
separating tetraalkylammonium salts from papulacandin
D. The authors then resolved to use tris(dimethylami-
no)sulfonium difluorotrimethylsilicate (TASF). In our
hands the reaction of compound 18 with TASF afforded
mainly pentaene 19 derived from elimination. We rea-
soned that TBAF could be employed if we could find the
correct conditions for chromatographic separation of the
product from the tetrabutylammonium salts. Thus after
evaluating a variety of solvents we were pleased to find
that column chromatography on silica gel under gravity
using a mixture of 5:40:55 of MeOH-MeCN-CH2Cl2 fur-
nished the final compound 1 in its pure form in 41%
yield.16
1H NMR (200 MHz, CDCl3): = 6.51 (dt, 1 H, J1,2 = 14.3 Hz,
J2,3 = 7.2 Hz, H-2), 5.95 (dt, 1 H, J1,2 = 14.3 Hz, J2,3 = 1.4 Hz, H-1),
2.04 (m, 2 H, H-3), 1.28 (m, 6 H, H-4,5,6), 1.28 (t, 3 H, J = 3.7, H-
7).
13C NMR (50 MHz, CDCl3): = 146.8 (C-2), 74.3 (C-1), 36.0
(CH2), 31.1 (CH2), 28.0 (CH2), 22.4 (CH2), 14.0 (CH3).
3-[(Triisopropylsilyl)oxy]propan-1-ol (5)
A mixture of propane-1,3-diol (42.77 g, 0.562 mol), TIPSCl (54.09
g, 0.281 mol), imidazole (85.82 g, 0.70 mol) in anhyd DMF (90 mL)
was stirred under N2 for 24 h at r.t. Et2O (600 mL) and aq sat. NH4Cl
solution (600 mL) were added. The layers were separated and the
organic layer was washed with aq sat. NH4Cl solution (3 300 mL)
and brine (600 mL), dried (MgSO4) and evaporated to give an oil.
The residue was purified by distillation (65–75 °C/0.1–0.2 Torr) to
Compound 1 was evaluated for its antifungal activity in
enzyme assay and cell based assays. Unfortunately, no
significant antifungal activity was found. This complete
loss of antifungal activity is surprising and seems to indi-
cate that the galactose moiety together with the short fatty
acid in saricandin is essential for activity. This is in con-
trast to the observation in the papulacandin. This finding give 43.6 g (67%) of 5.
is now under further investigation in our laboratory.
Synthesis 2001, No. 6, 833–840 ISSN 0039-7881 © Thieme Stuttgart · New York

PAPER
First Construction of a Saricandin Analog Corresponding to Papulacandin D
837
1H NMR (200 MHz, CDCl3): = 3.92 (t, 2 H, J = 5.5, H-3), 3.87–
3.79 (m, 2 H, H-1), 2.84–2.76 (m, 1 H, OH), 1.84–1.74 (m, 2 H, H-
2), 1.23–1.05 [m, 21 H, (CH3)2CHSi).
1H NMR (200 MHz, CDCl3): = 4.07 (t, 2 H, J = 6.3 Hz, CH2OSi),
2.77 (t, 2 H, J = 6.3 Hz, H-4), 1.05 (br s, 21 H, [CH3)2CHSi], 0.23
(s, 9 H, CH3Si).
13C NMR (50 MHz, CDCl3): = 63.7 (CH2OH), 62.8 (CH2OTIPS),
34.1 (CH2), 17.9 [(CH3)2CHSi], 11.8 [(CH3)2CHSi].
MS (EI+): m/z (%) = 189.2 (M+ – C3H7, 39.5), 157.2 (23.6), 131.1
13C NMR (50 MHz, CDCl3): = 186.4 (CO), 101.9 (C½CCO), 98.0
(SiC½C), 59.0 (CH2), 48.5 (CH2), 17.9 [(CH3)2CHSi], 12.5
[(CH3)2CHSi], -0.2 (CH3Si).
(73.7), 119.1 (100.0), 103.1 (65), 75.0 (73.7).
Anal. cald for C17H34O2Si2: C, 62.51; H, 10.49; found: C, 62.43; H,
10.56.
Anal. calcd for C12H28O2Si: C, 62.01; H, 12.14 ; found: C, 61.79; H,
12.01.
(S)-(-)-3-Hydroxy-5-[(triisopropylsilyl)oxy]-1-
trimethylsilylpent-1-yne [(-)-7]
(R,S)-3-Hydroxy-5-[(triisopropylsilyl)oxy]-1-
An oven-dried 250 mL round bottom-flask, equipped with a mag-
netic stirring bar, reflux condenser (with a septum) connected to a
schlenk line was purged with N2. 9-BBN (73.5 mL, 36.74 mmol, 0.5
M in THF) was added via cannula (through the condenser) followed
by ( )-( )-pinene (6.51 mL, 40.85 mmol, 9%, 97% ee). The solu-
tion was stirred at reflux for 3 h, then cooled to r.t. The ketone 8 (6.0
g, 18.37 mmol) dissolved in anhyd THF (15 mL) was tranferred to
the flask via a syringe. The solution turned to yellow. The mixture
was stirred at r.t. for 66 h. Acetaldehyde (1.9 mL) was added and
mixture was stirred for 15 min. The flask was kept in a water bath
and the solvent was removed by stirring under a stream of N2 for 5
h. The operation was completed by stirring the residue at 40 °C un-
der vacuum for an additional hour. The flask was filled with N2 and
anhyd Et2O (50 mL) was added. The solution was cooled to 0 °C
and ethanolamine (2.5 mL) was added. A white solid was formed.
The mixture was stirred for 15 min at 0 °C. The flask was then
opened to the air and the mixture was filtered and washed with cold,
anhyd Et2O (25 mL). The organic layer was washed with brine (300
mL), dried (MgSO4) and evaporated. The mixture was purified by
column chromatography (hexane Et2O, 98:2 to 95:5) to give 4.5 g
(75%) of the alcohol as a colorless oil in 97% ee as determined from
the Mosher esters; [ ]D23 –20.1 (c = 1.7, CHCl3).
trimethylsilylpent-1-yne [( )-7]
Step 1, 3-[(Triisopropylsilyl)oxy]propanal (6): A mixture of alcohol
5 (23.2 g, 0.01 mol) and PCC (32.3 g, 150 mmol) in anhyd CH2Cl2
(300 mL) was stirred under N2 at r.t. for 5 h. The solvent was re-
moved and Et2O (400 mL) was added. The mixture was stirred for
another 15 min and filtered through a pad of silica gel topped with
florisil. The residue was washed with Et2O. Removal of the solvent
under vacuo gave 19.9 g (87%) of pure aldehyde 6 as an oil.
1H NMR (200 MHz, CDCl3): = 9.82 (t, 1 H, J = 2.2 Hz, CHO),
4.08 (t, 2 H, J = 5.9 Hz, H-3), 2.61 (dt, 2 H, J = 5.9, 2.2 Hz, H-2),
1.29–0.98 [m, 21 H, (CH3)2CHSi].
13C NMR (50 MHz, CDCl3): = 202.2 (CHO), 57.8 (CH2O), 46.7
(CH2), 17.9 [(CH3)2CHSi], 11.8 [(CH3)2CHSi].
MS (EI+): m/z (%) = 229 (M+ – 1, 13.6), 187 (15.6), 145 (66.9), 119
(100.0), 103 (45.3,), 75 (55.7).
HRMS (EI+): m/z calcd for C12H25O2Si (M+- 1) 229.1624, found
229.1632.
Step 2, (R,S)-3-Hydroxy-5-[(triisopropylsilyl)oxy]-1-trimethylsi-
lylpent-1-yne [( )-7]: BuLi (1.98 M, 32.8 mL, 64.94 mmol) was
added to a solution of trimethylsilylacetylene (9.2 mL, 65.10 mmol)
in anhyd THF (50 mL) at 0 °C under N2. After stirring for 30 min at
the same temperature, the anion was transferred slowly via cannula
to a solution of the aldehyde 6 (10g, 43.4 mmol) in anhyd THF (100
mL) at 78 °C, under N2. The mixture was stirred at this tempera-
ture for 1 h and allowed to warm at r.t. The solvent was removed,
and Et2O (1 L) and aq sat. NH4Cl solution (1 L) were added. The
layers were separated and the organic layer was washed with aq sat.
NH4Cl solution (1L) and brine (1L), dried (MgSO4) and evaporated
to give an oil. The oil was purified by column chromatography (hex-
ane Et2O, 9:1) to give 10.5 g (74%) of the desired propargylic al-
cohol 7 as a yellow oil.
1H NMR (200 MHz, CDCl3): = 4.64 (dt, 1 H, JH,OH = 5.0, 6.5 Hz,
H-3), 4.21–4.13 (m, 1 H, H-5), 3.95–3.89 (m, 1 H, H-5), 3.64 ( d, 1
H, J = 5.0 Hz, OH), 2.08–1.98 (m, 1 H, H-4), 1.91–1.81 (m, 1 H, H-
4), 1.15–1.03 [m, 21 H, (CH3)2CHSi], 0.17 ( s, 9 H, CH3Si).
13C NMR (50 MHz, CDCl3): = 106.3 (C½CCHO), 88.9 (SiC½C),
62.0 (CHOH), 61.3 (CH2O), 38.5 (CH2), 17.8 [(CH3)2CHSi], 11.6
[(CH3)2CHSi], -0.2 (CH3Si).
1H NMR (200 MHz, CDCl3): = 4.64 (dt, 1 H, JH,OH = 4.0, 6.0 Hz,
H-3), 4.23–4.12 (m, 1 H, H-5), 4.00–3.87 (m, 1 H, H-5), 3.72 ( d, 1
H, J = 6.0 Hz, OH), 2.11–1.78 (m, 2 H, H-4), 1.14–1.03 [m, 21 H,
(CH3)2CHSi], 0.15 ( s, 9 H, CH3Si).
13C NMR (50 MHz, CDCl3): = 106.1 (C½CCO), 89.2 (SiC½C),
62.6 (CHOH), 61.7 (CH2O), 38.2 (CH2), 17.9 [(CH3)2CHSi], 11.7
[(CH3)2CHSi], -0.1 (CH3Si).
Anal. calcd for C17H36O2Si20.1H2O: C, 61.45; H, 10.05; found: C,
61.40; H, 10.17.
The asymmetric reduction was repeated on 16 g scale and the yield
was higher (82%).
(S)-( )-3-Hydroxy-5-[(triisopropylsilyl)oxy]pent-1-yne [( )-3]
The optically active alcohol ( )-7 (5.87g, 17.86 mmol) was dis-
solved in EtOH (90 mL). The solution was cooled to 0 °C in an ice-
bath. A solution of AgNO3 (12.14 g, 71.44 mmol) in a mixture of
EtOH H2O (390 mL v/v) was added dropwise through a funnel (45
min). A solid appeared immediately. After addition, the mixture
was stirred at the same temperature for 30 min, and then 60 min out-
side the ice-bath. A solution of KCN (16.2 g, 248.8 mmol) in H2O
(50 mL) was added and the mixture was stirred until the solid had
disappeared. More H2O was added (50 mL) and the mixture was ex-
tracted with hexane (3 175 mL). The organic layer was washed
with H2O, dried (MgSO4) and evaporated to give 4.18 g (92%) of
pure ( )-3 as an oil in 96% ee as determined from the Mosher esters;
[ ]D23 –19.2 (c = 0.47, CHCl3).
Anal. cald for C17H36O2Si2: C, 62.13; H, 11.04 ; found: C, 62.06; H,
11.10.
5-[(Triisopropylsilyl)oxy]-1-trimethylsilylpent-1-yn-3-one (8)
To a solution of the alcohol 7 (26.3 g, 80.03 mmol) in anhyd CH2Cl2
(400 mL) was added PCC (25.9 g, 120.15 mmol) in small portions.
The mixture was stirred under N2 at r.t. for 20 h. Solvent was re-
moved and Et2O (700 mL) was added. The suspension was filtered
through florisil and washed with Et2O (2 200 mL). The solvent
was removed under vacuo and the resulting oil was purified by col-
umn chromatography (hexane Et2O, 99:1) to give 18.6 g (71%) of
the desired ketone 8.
1H NMR (200 MHz, CDCl3): = 4.71–4.61 (m, 1 H, H-3), 4.23–
4.12 (m, 1 H, H-5), 3.98–3.88 (m, 1 H, H-5), 3.82 (d, 1 H, J = 6.1
Hz, OH), 2.47 (d, 1 H, J = 2.2, H-1), 2.13–1.80 (m, 2 H, H-4), 1.17–
1.00 [m, 21 H, (CH3)2CHSi].
Synthesis 2001, No. 6, 833–840 ISSN 0039-7881 © Thieme Stuttgart · New York

838
C. Hamdouchi, C. Sanchez-Martinez
PAPER
(4E,6E,3S)-( )-3-[(Triethylsily)oxy]-1-pivaloyldodeca-4,6-
13C NMR (50 MHz, CDCl3): = 84.4 (C-1), 72.8 (C-2), 62.1
(CHOH), 61.7 (CH2O), 38.2 (CH2), 17.9 [(CH3)2CHSi], 11.7
[(CH3)2CHSi].
diene [( )-(13)]
A solution of 11 (1.94 g, 9.78 mmol) and DMAP (1.31g, 10.76
mmol) in anhyd CH2Cl2 (35 mL) was cooled to 78 °C. Pivaloyl
chloride (1.19 mL, 9.70 mmol) was added via syringe very slowly
(15 min). The bath was removed and the mixture was allowed to
warm to r.t. for 3.5 h. DMAP (1.79 g, 16.67 mmol), imidazole
(1.0 g, 14.67 mmol), and TESCl (2.48 mL, 14.67 mmol) were added
sequentially and the mixture was stirred at the same temperature for
another 2.5 h. A solution of aq sat. NH4Cl (100 mL) was added and
the mixture was extracted with EtOAc (2 150 mL). The organic
layer was washed with brine and H2O, dried (MgSO4) and evaporat-
ed to give an oil that was purified by column chromatography (hex-
Anal. calcd for C14H28O2Si: C, 65.56; H, 11.00; found: C, 65.10; H,
10.88.
(6E,3S)-( )-3-Hydroxy-1-[(triisopropylsilyl)oxy]dodeca-6-en-
4-yne [( )-(10)]
Anhyd i-Pr2NH (4.32 mL, 30.8 mmol) and Pd(Ph3P)4 (0.42 g) were
added sequentially under a N2 stream to a stirred solution of the vi-
nyl iodide 2 (3.4 g, 15.2 mmol) dissolved in anhyd and degassed
benzene (62 mL) in 15 min. The reaction flask was protected against
light and the mixture was allowed to stir at r.t. for 45 min and then
a degassed solution of the terminal acetylene 3 (3.6 g, 14.04 mmol)
in anhyd benzene was added, followed by CuI (0.411 g). The mix-
ture was stirred in the dark at r.t. for 3.5 h. The reaction was trans-
ferred to a separating funnel and hexane (400 mL) and aq sat.
NH4Cl solution (400 mL) were added. The layers were separated
and aqueous layer was extracted with more hexane (200 mL). The
organic layers were combined and washed with aq sat. NH4Cl solu-
tion and brine, dried (MgSO4) and evaporated to give an oil. Purifi-
cation by column chromatography (hexane Et2O, 9:1) gave 3.9 g
23
ane EtOAc, 95:5) to provide 3.04 g (75%) of compound 13; [ ]D
–4 (c = 0.7, CHCl3).
1H NMR (200 MHz, CDCl3): = 6.15–5.91 (m, 2 H, H-5,6), 5.73–
5.45 (m, 2 H, H-4,7), 4.28–4.19 (m, 1 H, H-3), 4.11 (t, 2 H, J = 6.3
Hz, H-1), 2.11–2.01 (m, 2 H, H-8), 1.85–1.74 (m, 2 H, H-2), 1.45–
1.31 (m, 6 H, H-9,10,11), 1.19 (s, 9 H, t-C4H9), 0.94 (t, 9 H, J = 7.4
Hz, CH3CH2Si), 0.88 (t, 3 H, J = 6.8 Hz, H-12), 0.51 (m, 6 H,
CH3CH2Si).
13C NMR (50 MHz, CDCl3): = 178.3 (CO), 135.1 (CH=), 133.2
(CH=), 130.3 (CH=), 129.3 (CH=), 70.1 (CHOH), 61.0 (CH2O),
38.6 [C(CH3)3], 37.3 (CH2), 32.5 (CH2), 31.3 (CH2), 28.8 (CH2),
27.1 [C(CH3)3], 22.5 (CH2), 14.0 (CH3CH2), 6.7 (CH3CH2Si), 4.8
(CH3CH2Si).
23
(79%) of the coupling product 10 as an oil; [ ]D –19.6 (c = 0.48,
CHCl3).
1H NMR (200 MHz, CDCl3): = 6.12 (dt, 1 H, J = 7.0, 15.9 Hz, H-
7), 5.47 (dq, 1 H, J = 1.6, 15.9 Hz, H-6), 4.79–4.69 (m, 1 H, H-3),
4.20–4.09 (m, 1 H, H-1), 3.96–3.86 (m, 1 H, H-1), 3.68 ( d, 1 H,
J = 6.0 Hz, OH), 2.16–1.79 (m, 4 H, H-2,8), 1.40–1.20 (m, 6 H, H-
9,10,11), 1.35–0.89 [br s, 21 H, (CH3)2CHSi], 0.87 (t, 3 H, J = 6.7
Hz, H-12).
13C NMR (50 MHz, CDCl3): = 145.2 (CH=), 108.9 (CH=), 87.9
(C½C), 86.3 (C½C), 62.6 (CHOH), 61.7 (CH2O), 38.6 (CH2), 33.0
(CH2), 31.2 (CH2), 28.3 (CH2), 22.4 (CH2), 17.9 [(CH3)2CHSi], 14.0
(CH3CH2), 11.7 [(CH3)2CHSi].
MS (EI+): m/z (%) = 396.3 (M+, 0.8), 295.3 (11.3), 294.3 (35.2),
281.2 (17.8), 267.2 (37.4), 265.2 (19.8), 224.2 (27.7), 223.2 91.6),
57.1 (100.0).
Anal. calcd for C23H44O3Si: C, 69.64; H, 11.18; found: C, 69.54; H,
11.02.
(4E,6E,3S)-( )-3-[(Triethylsily)oxy]-1-hydroxydodeca-4,6-
diene [( )-14]
MS (FAB+): m/z = 353.2 ([MH]+).
HRMS (FAB): m/z calcd for C21H41O2Si (MH+) 353.2876, found
A stirred solution of the ester 13 (1.6 g, 4.03 mmol) in anhyd THF
(25 mL) under N2 was cooled to 78 °C. A solution of DIBAL
(14.12 mL, 1 M in hexane) was added dropwise (30 min). The mix-
ture was stirred for 2 h at the same temperature then quenched with
MeOH (0.5 mL) and transferred to a decanting funnel. A saturated
solution of sodium tartrate (150 mL) and EtOAc (600 mL) was add-
ed. After vigorous shaking the layers was separated and the organic
layer was washed with more sodium tartrate, dried (MgSO4) and
evaporated to give an oil. The oil was purified by column chroma-
tography (hexane Et2O, 97:3 to 90:10) to give 0.945 g (75%) of 14;
[ ]D23 –27 (c = 0.2, CHCl3).
353.2868.
Anal calcd for C21H40O2Si0.4H2O: C, 70.09; H, 11.43; found C,
70.16; H, 11.24.
(4E,6E,3S)-( )-Dodeca-4,6-dien-1,3-diol [( )-(11)]
LiAlH4 (36.9 mL, 1 M in Et2O) was added dropwise to a solution of
10 (4.34 g, 12.3 mmol) in anhyd THF (25 mL) at 0 °C under N2. Af-
ter addition, the ice-bath was removed and the mixture was stirred
at r.t. for 5 h. Et2O (25 mL) was added and the resulting solution
was cooled to 0 °C (ice-bath) before the hydrolysis. H2O (1.44 mL),
NaOH (15%, 1.44 mL) and H2O (4.32 mL) were added sequentially
dropwise with vigorous stirring. The solid was filtered and washed
with Et2O. The solvent was removed and the oil was purified by col-
umn chromatography (hexane EtOAc, 6:4) to afford 1.89 g (78%)
of the desired diol 11; [ ]D23 –11.3 (c = 0.31, CHCl3).
1H NMR (200 MHz, CDCl3): = 6.19–5.93 (m, 2 H, H-5,6), 5.68
(dt, 1 H, J7,8 = 6.8, J6,7 = 14.6 Hz, H-7), 5.56 (dd, 1 H, J3,4 = 6.8,
J4,5 = 14.6 Hz, H-4), 4.47–4.38 (m, 1 H, H-3), 3.89–3.64 (m, 2 H, H-
1), 2.66 (t, 1 H, J = 5.3 Hz, OH), 2.12–2.02 (m, 2 H, H-8), 1.90–1.66
(m, 2 H, H-2), 1.46–1.24 (m, 6 H, H-9,10,11), 1.00–0.91 (m, 9 H,
CH3CH2Si), 0.88 (t, 3 H, J = 6.8 Hz, H-12), 0.66–0.54 (m, 6 H,
CH3CH2Si).
13C NMR (50 MHz, CDCl3): = 135.4 (CH=), 132.8 (CH=), 130.4
(CH=), 129.2 (CH=), 73.2 (CHO), 60.5 (CH2OH), 39.6 (CH2),
32.6 (CH2), 31.4 (CH2), 28.8 (CH2), 22.5 (CH2), 14.0 (CH3CH2), 6.7
(CH3CH2Si), 4.8 (CH3CH2Si).
1H NMR (200 MHz, CDCl3): = 6.23 (ddd, 1 H, J4,5 = 15.0,
J5,6 = 10.3 Hz, J5,3 = 1.1 Hz, H-5), 6.02 (ddt, 1 H, J6,7 = 14.8,
J5,6 = 10.2, J6,8 = 1.3 Hz, H-6), 5.71 (dt, 1 H, J7,8 = 6.8, J6,7 = 15.0 Hz,
H-7), 5.62 (dd, 1 H, J7,8 = 6.7, J3,4 = 15.0 Hz, H-4), 4.46–4.36 (m, 1
H, H-3), 3.89–3.79 (m, 2 H, H-1), 2.24 (t, 1 H, J = 4.9 Hz, OH), 2.20
(d, 1 H, J = 3.4 Hz, OH), 2.12–2.02 (m, 2 H, H-8), 1.84–1.75 (m, 2
H, H-2), 1.45–1.19 (m, 6 H, H-9,10,11), 0.88 (t, 3 H, J = 6.6 Hz, H-
12).
Anal. calcd for C18H36O2Si: C, 69.17; H, 11.61; found: C, 69.26; H,
11.57.
(4E,6E,3S)-( )-3-[(Triethylsily)oxy]dodeca-4,6-dienal [( )-15]
TPAP (5.1 mg, 0.0145 mmol, 5%) was added in one portion to a
stirred mixture of the alcohol 14 (0.09 g, 0.29 mmol), NMO (0.051
g, 0.435 mmol) and freshly activated powered molecular sieves
(150 mg) in anhyd CH2Cl2 (1.5 mL) under N2. After 30 min, more
CH2Cl2 was added and filtered trough a florisil-Celite pad. This pad
13C NMR (50 MHz, CDCl3): = 135.6 (CH=), 132.7 (CH=), 130.6
(CH=), 129.2 (CH=), 71.6 (CHOH), 60.4 (CH2OH), 38.5 (CH2),
32.5 (CH2), 31.3 (CH2), 28.8 (CH2), 22.4 (CH2), 13.9 (CH3CH2).
Anal. calcd for C12H22O20.75H2O: C, 68.04; H, 11.19; found: C,
68.16; H, 10.35.
Synthesis 2001, No. 6, 833–840 ISSN 0039-7881 © Thieme Stuttgart · New York

PAPER
First Construction of a Saricandin Analog Corresponding to Papulacandin D
839
was washed with EtOAc (25 mL). Solvents were removed and the
residue was purified by column chromatography (hexane-Et2O,
95:5) to give 28 mg (76%) of 15; [ ]D23 –19.1 (c = 1.15, CHCl3).
1H NMR (200 MHz, CDCl3): = 7.34 (dd, 1 H, J2,3 = 15.4,
J3,4 = 10.0 Hz, H-3), 6.33–5.92 (m, 4 H, H-4,5,9,10), 5.80 (d, 1 H,
2,3 = 15.4 Hz, H-2), 5.67 (dt, 1 H, J11,12 = 6.9, J10,11 = 14.4, H-11),
J
5.51 (dd, 1 H, J7,8 = 6.8, J8,9 = 14.4 Hz, H-8), 4.32–4.14 (m, 1 H, H-
7), 2.42–2.36 (m, 2 H, H-6), 2.13–2.01 (m, 2 H, H-12), 1.41–1.25
(m, 6 H, H-13,14,15), 0.97–0.89 (m, 9 H, CH3CH2Si), 0.88 (t, 3 H,
J = 6.8 Hz, H-16), 0.63–0.55 (m, 6 H, CH3CH2Si).
1H NMR (200 MHz, CDCl3): = 9.77 (t, 1 H, J = 2.3 Hz, CHO),
6.17 (ddd,, 1 H, J3,5 = 0.9, J4,5 = 14.9, J5,6 = 10.2 Hz, H-5), 6.00 (ddt,
1 H, J6,8 = 1.2, J5,6 = 10.3, J6,7 = 14.7 Hz, H-6), 5.82 (dt, 1 H,
J7,8 = 7.3, J6,7 = 14.7 Hz, H-7), 5.56 (dd, 1 H, J3,4 = 6.7, J4,5 = 14.9 Hz,
H-4), 4.72–4.62 (m, 1 H, H-3), 2.70–2.45 (m, 2 H, H-2), 2.12–2.02
(m, 2 H, H-8), 1.54–1.23 (m, 6 H, H-9,10,11), 0.97–0.91 (m, 9 H,
CH3CH2Si), 0.88 (t, 3 H, J = 6.8 Hz, H-12), 0.64–0.52 (m, 6 H,
CH3CH2Si).
13C NMR (50 MHz, CDCl3): = 201.8 (CHO), 136.1 (CH=), 132.0
(CH=), 130.7 (CH=), 129.0 (CH=), 68.9 (CHO), 51.7 (CH2CHO),
32.6 (CH2), 31.4 (CH2), 28.8 (CH2), 22.5 (CH2), 14.0 (CH3CH2), 6.7
(CH3CH2Si), 4.8 (CH3CH2Si).
1,1-Anhydro-1-C-[6-(hydroxymethyl)-2,4-
bis[(triisopropylsilyl)oxy]phenyl]-3-O-[(2E,4E,8E,10E,7S)-7-
[(triethoxysilyl)oxy]hexadeca-2,4,8,10-tetraenoyl]-4,6-O-(di-
tert-butylsilylene)-b-D-glucopyranose (C-3 18)
A mixture of acid 17 (60 mg, 0.16 mmol), Et3N (40.1 L, 0.288
mmol), 2,4,6-trichlorobenzoyl chloride (30.5 L, 0.195 mmol) and
DMAP (8.5 mg), under N2, in anhyd THF (2 mL) was allowed to
react at r.t. for 1 h in dark (aluminum foil). The solvent was re-
moved under high vacuum and the remaining mixed anhydride was
dissolved in anhyd DMF (1.7 mL) and transferred via cannula to a
mixture of protected spiroketal 4 (0.18 g, 0.24 mmol) and DMAP
(48.9 mg) in anhyd DMF (2.1 mL) at r.t. After 5 h, H2O was added
and the mixture was extracted with EtOAc (2 25 mL). The organic
layer was washed with more H2O (2 times), brine (2 times), dried
(MgSO4) and evaporated to give a 3:1 mixture of the C-3 and C-2
MS (FAB+): m/z = 310.3 (M+), 281.2 (M+ – CHO).
Anal. calcd for C18H34O2Si: C, 69.61; H, 11.03; found: C,69.27; H,
10.83.
Ethyl (2E,4E,8E,10E,7S)-(+)-7-[(Triethylsily)oxy]hexadeca-
2,4,8,10-tetraenoate [(+)-16]
LiOBu-t (0.35 mmol, 0.35 mL, 1 M in THF) was added to a cooled
( 78 °C) solution of triethyl phosphonocrotonate (9; 0.11 mL, 0.48
mmol) under N2. The dry ice bath was removed and the brown so-
lution was stirred for 15 min, then recooled to 78 °C. The solution
was transferred via cannula to a cooled ( 78 °C) solution, under N2,
of aldehyde 15 (50 mg, 0.16 mmol) in anhyd THF (3 mL). The re-
sulting mixture was stirred at the same temperature for 30 min. The
dry ice-bath was replaced by an ice-bath and the solution was stirred
for another 2 h. Et2O (40 mL) and aq sat. NH4Cl solution H2O (v/
v, 20 mL) were added. The layers were separated and the organic
layer was washed with more H2O, dried (Na2SO4) and evaporated
to give a yellow oil. Purification by column chromatography (hex-
mixture. The mixture was purified by chromatography (hexane-
23
EtOAc, 95:5) to afford 55 mg (31%) of 18 as a white solid; [ ]D
8.4 (c = 0.95, CHCl3).
–
1H NMR (200 MHz, CDCl3): = 7.28 (dd, 1 H, J3,4 = 10.3,
2,3 = 15.3 Hz, H-3'), 6.31 (d, 1 H, J = 1.8, ArH), 6.23 (d, 1 H, J = 1.8,
J
ArH), 6.29–5.97 (m, 4 H, H-4,' 5',9',10'), 5.89 (d, 1 H, J = 15.3 Hz,
H-2'), 5.66 (dt, J11,12 = 6.9, J10,11 = 14.5 Hz, H-11'), 5.51 (dd, 1 H,
J7,8 = 6.8, J8 9 = 14.5 Hz, H-8'), 5.30 (t, 1 H, J = 9.3 Hz, H-3), 5.16 (d,
1 H, J = 12.9 Hz, ABsystem), 5.02 (d, 1 H, J = 12.9 Hz, ABsystem),
4.40 (t, J = 10.3 Hz, H-2), 4.22–3.77 (m, 5 H, sugar + H-7'), 2.40–
21.34 (m, 2 H, H-6'), 2.12–1.90 (m, 2 H, H-12'), 1.38–1.18 (m, 6 H,
H-13',14',15'), 1.15–0.89 (m, 69 H, t-C4H9Si, CH3CH2Si, i-C3H7),
0.88 (t, 3 H, J = 6.4 Hz, H-16'), 0.63–0.50 (m, 6 H, CH3CH2Si).
23
ane Et2O, 95:5) gave 48 mg (74%) of 16; [ ]D +6.1 (c = 0.64,
CHCl3).
13C NMR (50 MHz, CDCl3): = 167.5, 158.7, 152.1, 144.9, 143.2,
140.1, 135.3, 133.2, 130.6, 130.3, 129.3, 119.8, 119.1, 111.0, 109.5,
104.8, 74.8, 73.3, 72.7, 71.5, 68.7, 66.8, 42.2, 32.6, 31.4, 29.7, 28.9,
27.3, 27.2, 22.7, 22.5, 19.9, 18.0, 17.8, 14.0, 13.2, 12.6, 6.8, 4.9.
1H NMR (200 MHz, CDCl3): = 7.25 (dd, 1 H, J2,3 = 15.4,
J3,4 = 10.0 Hz, H-3), 6.26–5.91 (m, 4 H, H-4,5,9,10), 5.78 (d, 1 H,
J2,3 = 15.4 Hz, H-2), 5.66 (dt, 1 H, J11,12 = 7.1, J10,11 = 14.6 Hz, H-
11), 5.50 (dd, 1 H, J7 8 = 6.8, J8 9 = 14.6 Hz, H-8), 4.26–4.13 (m, 1
H, H-7), 4.20 (q, 2 H, J = 7.1 Hz, CH3CH2O), 2.40–2.33 (m, 2 H, H-
6), 2.12–2.01 (m, 2 H, H-12), 1.45–1.19 (m, 6 H, H-13,14,15), 1.29
(t, 3 H, J = 7.1, CH3CH2O), 0.97–0.89 (m, 9 H, CH3CH2Si), 0.88 (t,
3 H, J = 6.8 Hz, H-16), 0.63–0.50 (m, 6 H, CH3CH2Si).
13C NMR (50 MHz, CDCl3): = 167.3 (CO), 144.8 (CH=), 140.3
(CH=), 133.5 (CH=), 133.1 (CH=), 130.5 (CH=), 130.4 (CH=),
129.3 (CH=), 119.7 (CH=), 72.6 (CHO), 60.2 (CH3CH2O), 42.2
(CH2), 32.6 (CH2), 31.4 (CH2), 28.9 (CH2), 22.5 (CH2), 14.3
(CH3CH2O), 14.0 (CH3CH2), 6.8 (CH3CH2Si), 4.9 (CH3CH2Si).
Anal. calcd for C61H108O10Si4: C, 65.78; H, 9.77; found: C, 66.03;
H, 9.86.
1,1-Anhydro-1-C-[6-(hydroxymethyl)-2,4-dihydroxyphenyl]-3-
O-[hexadeca-2E,4E,8E,10E-tetraenoyl]-b-D-glucopyranose (1)
TBAF (115 L, 0.115 mmol, 1 M in THF) was added in one portion
to a solution of the ester 18 (20 mg, 0.018 mmol) in anhyd THF
(1 mL) at r.t., under N2, in dark (aluminum foil). The mixture was
stirred at the same temperature for 3.5 h. A drop of MeOH was add-
ed and solvents were removed under vacuo. The resulting crude was
immediately purified by column chromatography (MeOH MeCN
CH2Cl2, 5:45:55). As it was pointed out in the text after an exhaus-
tive study the mixture of MeOH MeCN CH2Cl2 was found to be
effective for the purification of the final product 1 from the tetrabu-
tylammonium salts. Additionally the column chromatography
should be protected against light and performed without pressure.
Following these conditions the desired compound 1 was isolated in
its pure form; yield: 4 mg (41%); [ ]D23 +25 (c = 0.1, CHCl3).
1H NMR (200 MHz, CD3OD): = 7.32 (dd, 1 H, J3,4 = 10.3,
J2,3 = 15.3 Hz, H-3'), 6.39–5.85 (m, 5 H, H-2',4',5',9',10'), 6.18 (br s,
2 H, ArH), 5.68 (dt, 1 H, J11,12 = 7.0, J10,11 = 14.7 Hz, H-11'), 5.56
(dd, 1 H, J7,8 = 6.7, J8,9 = 14.5 Hz, H-8'), 5.34 (t, 1 H, J = 9.6 Hz, H-
3), 5.07 (d, 1 H, J = 12.6 Hz, ABsystem), 4.98 (d, 1 H, J = 12.6 Hz,
ABsystem), 4.33 (d, 1 H, J = 10.0 Hz, H-2), 4.21–4.11 (m, 1 H, H-
7'), 3.92–3.63 (m, 4 H, sugar), 2.40 (m, 2 H, H-6'), 2.12–2.01 (m, 2
Anal. calcd for C24H42O3Si: C, 70.88; H, 10.41; found: C, 71.01; H,
10.48.
(2E,4E,8E,10E,7S)-7-[(Triethylsily)oxy]hexadeca-2,4,8,10-
tetraenoic Acid (17)
KOSiMe3 (205 mg, 1.6 mmol) was added in one portion to a solu-
tion of ester 16 (65 mg, 0.16 mmol) in anhyd THF (3.5 mL) at r.t.
The mixture was stirred at the same temperature for 3.5 h. A solu-
tion of citric acid (3.4 mL, 0.327 g in H2O) was added and the mix-
ture was extracted with EtOAc (3 10 mL). The organic layer was
washed with brine, dried (MgSO4) and evaporated to give 17 as a
yellow solid in 91% yield that was used in the next step without pu-
rification. It should be noted that all attempts to purify the acid by
column chromatography had led to decomposition. The compound
should be used shortly after its preparation because of its instability.
Synthesis 2001, No. 6, 833–840 ISSN 0039-7881 © Thieme Stuttgart · New York

840
C. Hamdouchi, C. Sanchez-Martinez
PAPER
(8) The (E)-iodoheptene was prepared from the correspoding
H, H-11'), 1.46–1.28 (m, 6 H, H-13',14'; 5'), 0.89 (t, 3 H, J = 6.9 Hz,
H-16').
heptyne according to the procedure described by Zweifel
and co-workers: Zweifel, G.; Whitney, C. C. J. Am. Chem.
Soc. 1967, 89, 2753.
Anal. calcd for C29H38O10: C, 63.72; H, 7.01; found: C, 63.60; H,
7.07.
(9) Sonogashira, K.; Tohda, Y.; Hagihara, N. Tetrahedron Lett.
1975, 4467.
(10) Rossi, R.; Carpita, A. Synthesis 1977, 561.
(11) Solladie, G.; Hamdouchi, C. Synlett 1989, 66.
(12) Dess, D. B.; Martin, J. C. J. Org. Chem. 1983, 48, 4155.
(13) Ireland, R. E.; Wardle, R. B. J. Org. Chem. 1987, 52, 1780.
(14) For an excellent review, see: Ley, S. V.; Norman, J.; Griffith,
W. P.; Marsden, P. S. Synthesis 1994, 639.
(15) The reason of this selection is supported by the fact that
Barrett’s approach is the only known efficient approach that
delivers directly the spiroketal nucleus bearing the
protecting group of choice for the generation of the correct
monoester. On the other hand their study on protecting
groups demonstrated that the protecting group of choice for
this type of systems are O-4”,O-6”-di-tert-butylsilylene, and
phenol tris(triisopropylsilyl)
Acknowledgement
This research was supported by a CDTI program (Plan concertado
94/0036) and the Spanish PROFARMA programme (Ministerio de
Industria and Ministerio de Sanidad). C. S. is grateful to the Mini-
stery of education for a fellowship. We are also grateful to Dr. Al-
mudena Rubio and Dr. Steve Hitchcock for useful suggestions.
References and Notes
(1) (a) Traxler, P.; Gruner, J.; Auden, J. A. L. J. Antibiot. 1977,
30, 289. (b) Traxler, P.; Frittz, H.; Richter, W. J. Helv. Chim.
Acta 1977, 60, 578. (c) Traxler, P.; Fritz, H.; Richter, W. J.;
Fuhrer, H. J. Antibiot. 1980, 33, 967.
(2) (a) Danishefsky, S.; Phillips, G.; Ciufolini, M. Carbohydr.
Res. 1987, 171, 317. (b) Friesen, R. W.; Sturino, C. F. J.
Org. Chem. 1990, 55, 5808. (c) Dubois, E.; Beau, J. M.
Carbohydr. Res. 1992, 223, 157. (d) Dubois, E.; Beau, J. M.
Tetrahedron Lett. 1990, 31, 5165. (e) Czernecki, S.; Perlat,
M. C. J. Org. Chem. 1991, 56, 6289. (f) Schmidt, R. R.;
Frick, W. Tetrahedron 1988, 44, 7163. (g) Barrett, A. G.
M.; Pena, M.; Willardsen, J. A. J. Chem. Soc., Chem.
Commun. 1995, 1147.
(3) Barett, A. G. M.; Pena, M.; Willardsen, J. A. J. Org. Chem.
1996, 61, 1082.
(4) Chen, R. H. J. Antibiot. 1996, 49, 596.
(5) The analogous papulacandin D is known as the least
abundant of papulacandins family. Compound 1 has not
been isolated, probably because of the extreme scarcity of
the natural material.
HO
OH
1) TIPSCl, DMAP, DMF, Im
2) LiAlH4, Et2O
TIPSO
OTIPS
Br
3) NBS, CCl4
4) TBDMSCl, DMAP, DMF, Im
CO2Me
TBSO
1) t-BuLi, Et2O
OTIPS
O
O
TMSO
TMSO
OTMS
TIPSO
O
OTMS
O
O
t-Bu
Si
2) Amberlite IR-120 (Plus), MeOH
23 °C, 16h
3) (t-Bu)2Si(OTf)2, 2,6-lutidine, CH2Cl2
O
OH
t-Bu
OH
(6) The racemic building block ( )-7 was initially converted into
a 1:1 mixture of 1 and its C-7 epimer following the same
sequences employed for the preparation of 1 in
enatiomrically pure form.
(7) Midlandess, M. M.; Mc Dowell, D. C.; Hatch, R. L.;
Tramontano, A. J. Org. Chem. 1980, 102, 867.
(16) The mixture of MeOH CH3CN CH2Cl2 solvents turned out
to be very useful and was applied successfully for the
purification of other polar compounds from
tetrabutylammonium salts and other salts that derived from
Wittig or Horner-Emmons reactions.
Synthesis 2001, No. 6, 833–840 ISSN 0039-7881 © Thieme Stuttgart · New York